Visualization technologies developed at Uber are crtically important to understanding the decisions made by an autonomous vehicle. The Data Visualization team describes how they built the web-based platform and the challenges of combining complex and diverse datasets into a reusable web component. One of the challenges is that a large amount of data is required to recreate the context around a trip, like preprocessed maps and runtime vehicle logs. The mind-boggling video below shows the detail required.
Vehicle logs describe what the vehicle is doing, what it sees, and how it acts. Algorithms are run based on sensor data: perception (measuring), prediction (forecasting), and motion planning (acting). A vehicle needs to be able to perceive the activity around it through sensors and predict where these objects will be in the near future.
There’s much more, including a pitch to join the team if you have an interest in mapping self-driving technologies. Read the whole thing – fascinating.